San Diego R Users Group

I’m the organizer for the R user group in San Diego. We’re still a bit new so we haven’t had a lot of meetings but I try to do what I can. If you have any tips or would like to give a talk please let me know! I started this group so San Diego R users can get together, learn from each other, and talk about all the great things R has to offer. R has a fantastic online community and I want to help establish that at a local level.

Seefood Shiny App

I made a Shiny application based on the Seefood app from HBO’s Silicon Valley. Given an image, it predicts whether the image is a hot dog or ‘not hot dog’. I built a Keras model on Paperspace and wrapped it in a Shiny application. You can find the app here and the modeling code here.

R package geblm

I have an R package called geblm that implements some research of a previous professor. It implements MCMC algorithms for certain Bayesian linear and liner mixed models that have geometric convergence rates. It’s kind of a small area of research with not a lot of results yet so the use case is pretty limited. It still needs some work but unfortunately I just haven’t had the time after the semester I worked on it. I did learn a lot, however, like using C++ with Rcpp, writing unit tests, and generally how R packages work. This experience served as a foundation I used later when building R packages at work.

R Workshops at SDSU

I did a few workshops for a student organization, the Society of Statisticians and Actuaries, when I was a student at SDSU. I gave workshops on an introduction to R, data wrangling with dplyr, data visualization with ggplot2, and an introduction to Git & GitHub. Materials can be found here. This was a great experience because it forced me to think hard about what I feel is important about each topic/tool and why others should care.

Adventures In Bayesian Structural Time Series

I did a group project for a Bayesian statisics course where we were tasked with making video tutorials on a topic of our choosing. We talked about the theory behind Bayesian structural time series models and showed how to implement the theory with the bsts package. We had a ton of fun with this project! The videos had a DnD/adventure theme where viewers were on quest to learn about the topic (we may have went a little overboard with this theme 😄). The videos can be found here and all the materials can be found here.